Glass prime coat detection method and application equipment thereof
Through image acquisition and area classification technology, convolutional neural networks are used to detect glass primers, which solves the problem of insufficient detection accuracy in existing technologies, realizes efficient and accurate primer detection, and improves production efficiency.
Patent Information
- Application Number
- CN202510824763.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-19
- Publication Date
- 2025-09-26
Smart Images
Figure CN120707526A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of glass production detection, and in particular to a glass primer detection method, a glass primer detection device, a glass primer detection equipment, a glass primer system and a computer-readable storage medium. Background Art
[0002] To protect the glass, it is often primed during the glass production process. This involves spraying a layer of primer that matches the glass surface to create a uniform base color. This primer not only serves as a base for color coatings or other exterior effect coatings, but also enhances the surface hardness of the material, protects against moisture and corrosion, and other effects. Because parameters such as the primer range and thickness have a significant impact on the subsequent assembly process, primer testing is necessary. However, in related technologies, grayscale testing is generally used to test primed glass, which lacks accuracy. Summary of the Invention
[0003] In view of the above problems, embodiments of the present invention are proposed to provide a glass primer detection method, a glass primer detection device, a glass primer detection equipment, a glass primer system and a computer-readable storage medium that overcome the above problems or at least partially solve the above problems.
[0004] In order to solve the above problems, in a first aspect of the present invention, an embodiment of the present invention discloses a glass primer detection method, comprising: During the glass primer coating process, a glass primer coating image is collected; Performing region classification on the glass primer image to determine at least one primer region; The primer area is compared with a preset detection template, and primer qualified information is generated when all the primer areas match the preset detection template.
[0005] Optionally, the method further includes: In the case where at least one of the primer areas does not match the preset detection template, primer failure information is generated.
[0006] Optionally, the method further includes: Obtain annotated sample sets and initial models; The initial model is trained based on the labeled sample set until the training confidence of the initial model is greater than a preset confidence threshold, and a classification model is generated. The classification model is used to perform regional classification of the glass primer image and determine at least one primer area.
[0007] Optionally, the step of training the initial model based on the labeled sample set includes: The labeled sample set is input into the initial model for training based on a preset hyperbolic tangent function.
[0008] Optionally, the method further includes: The glass primer image is saved in the annotated sample set.
[0009] Optionally, the preset detection template includes a coverage area, and the comparing the primer area with the preset detection template includes: comparing the primer area to the covered area; In a case where the primer area is identical to the coverage area, determining that the primer area matches a preset detection template; In a case where the primer area is different from the covering area, it is determined that the primer area does not match the preset detection template.
[0010] In a second aspect of the present invention, an embodiment of the present invention discloses a glass primer detection device, comprising: A first acquisition module is used to collect a glass primer image during the glass primer coating process; a classification module, configured to perform regional classification on the glass primer image to determine at least one primer region; The comparison module is used to compare the primer area with a preset detection template, and generate primer qualified information when all the primer areas match the preset detection template.
[0011] In the third aspect of the present invention, an embodiment of the present invention discloses a glass primer detection device, including a processor, a memory, and a computer program stored in the memory and capable of running on the processor, and when the computer program is executed by the processor, the steps of the glass primer detection method as described above are implemented.
[0012] In the fourth aspect of the present invention, an embodiment of the present invention discloses a glass primer system, a robot, a glass primer process equipment and the glass primer detection equipment as described above, the glass primer process equipment and the glass primer detection equipment are installed on the robot; the robot drives the glass primer process equipment and the glass primer detection equipment to move; the glass primer process equipment is used to perform primer process treatment on the glass.
[0013] In a fifth aspect of the present invention, an embodiment of the present invention discloses a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the glass primer detection method described above are implemented.
[0014] The embodiments of the present invention include the following advantages: The embodiment of the present invention collects a glass primer image during the glass primer coating process; performs regional classification on the glass primer image to determine at least one primer area; compares the primer area with a preset detection template, and generates primer qualification information when all the primer areas match the preset detection template. By collecting the glass primer image during the primer coating process and performing detection based on it, the primer condition of the glass can be detected in a timely manner, which improves the timeliness of the detection, so that primer defects can be repaired in a timely manner and improves production efficiency. The primer area and the background area are distinguished in the glass primer image, and then the actual glass primer condition is identified from the primer area based on the comparison with the preset detection template to determine whether the glass primer is qualified. This can reduce the situation where detection errors are caused by the glass and the background color being similar, thereby improving the accuracy of detection. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 This is a flowchart of a glass primer detection method embodiment of the present invention; Figure 2 is a flowchart of another embodiment of a glass primer detection method of the present invention; Figure 3 This is a schematic diagram of a glass primer test result of the present invention; Figure 4 Schematic diagram of glass primer failure detection of the present invention; Figure 5 This is a flowchart of an example of a glass primer detection method of the present invention; Figure 6 This is a structural block diagram of an embodiment of a glass primer detection device of the present invention; Figure 7 This is a structural block diagram of a glass primer detection device provided by the present invention; Figure 8 This is a structural block diagram of a glass primer system provided by the present invention; Figure 9 This is a structural block diagram of a storage medium provided by the present invention. DETAILED DESCRIPTION
[0016] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0017] Reference Figure 1 , shows a flowchart of a glass primer detection method embodiment of the present invention, which may specifically include the following steps: Step 101, during the glass primer coating process, collecting a glass primer coating image; In a device that performs a primer coating process on glass, an image of the glass surface after the primer coating process, namely, a glass primer coating image, can be collected.
[0018] Step 102, performing region classification on the glass primer image to determine at least one primer region; The glass primer image can be regionally classified to distinguish areas that have undergone primer coating from areas that have not undergone primer coating, thereby determining at least one primer region. The primer region is the glass surface area that has undergone primer coating. The area that has not undergone primer coating is the background area.
[0019] Step 103 : comparing the primer area with a preset detection template, and generating primer qualified information when all the primer areas match the preset detection template.
[0020] After obtaining the primer area, each primer area can be compared with a preset detection template to determine whether the current primer area is qualified. After all primer areas are compared with the preset detection template, if all primer areas match the preset detection template, primer qualified information can be generated. The primer qualified information indicates that the current glass primer process is qualified, and subsequent process processing can be performed to ensure the performance of the glass. Among them, the primer parameters that the preset detection template indicates are qualified can be formulated in advance according to the process parameters in the product prefabrication process of the primer process. The specific content can be determined according to the actual primer process, and the embodiment of the present invention does not specifically limit this.
[0021] The embodiment of the present invention collects a glass primer image during the glass primer coating process; performs regional classification on the glass primer image to determine at least one primer area; compares the primer area with a preset detection template, and generates primer qualification information when all the primer areas match the preset detection template. By collecting the glass primer image during the primer coating process and performing detection based on it, the primer condition of the glass can be detected in a timely manner, which improves the timeliness of the detection, so that primer defects can be repaired in a timely manner and improves production efficiency. The primer area and the background area are distinguished in the glass primer image, and then the actual glass primer condition is identified from the primer area based on the comparison with the preset detection template to determine whether the glass primer is qualified. This can reduce the situation where detection errors are caused by the glass and the background color being similar, thereby improving the accuracy of detection.
[0022] Reference Figure 2 , shows a flowchart of another embodiment of a glass primer detection method of the present invention, which may specifically include the following steps: Step 201, during the glass primer coating process, collecting a glass primer coating image; In an embodiment of the present invention, a visual sensor such as a camera can be set to collect images. The visual sensor can move with the equipment performing the primer process and can collect the glass primer image in time during the glass primer process.
[0023] Step 202, performing region classification on the glass primer image to determine at least one primer region; The pixel areas in the glass primer image may be classified, and different parts in the glass primer image may be divided, and at least one primer area may be determined from these areas.
[0024] In an optional embodiment of the present invention, the method further includes: Step S1, obtaining a labeled sample set and an initial model; A labeled sample set that has been pre-labeled for the region and an initial model for classification can be obtained. The labeled sample set includes multiple sample data, and the image in each sample data is labeled with a specific background area and primer area. The initial model can be a multi-layer perceptron (MLP), autoencoder (Autoencoder), convolutional neural network (CNN), recurrent neural network (RNN), long short-term memory network (LSTM), gated recurrent unit (GRU), graph neural network (GNN), and capsule network, etc. The embodiment of the present invention does not specifically limit the type of initial model.
[0025] Step S2: training the initial model based on the labeled sample set until the training confidence of the initial model is greater than a preset confidence threshold, generating a classification model, and the classification model is used to perform regional classification of the glass primer image and determine at least one primer area.
[0026] Each labeled sample in the labeled sample set can be used to train the initial model accordingly. After each training session, the training confidence of the current initial model is determined. When the training confidence of the initial model is greater than a preset confidence threshold, it indicates that the current level of training meets the usage requirements. The model of the current training round is determined as a classification model. The preset confidence threshold can be determined based on actual conditions and is not specifically limited in this embodiment of the present invention.
[0027] Specifically, the step of training the initial model based on the labeled sample set includes: inputting the labeled sample set into the initial model for training based on a preset hyperbolic tangent function.
[0028] In an embodiment of the present invention, for model training, a preset hyperbolic tangent function may be used as an activation function of the initial model, and each sample in the labeled sample set may be input into the initial model for training.
[0029] Taking convolutional neural networks as an example, a multi-dimensional dynamic feature recognition method is used to identify the primer and background. Using the primer sample image collected in pre-debugging, the "primer" and "background" areas are manually selected according to the standard. Then, a custom calculation formula is used to train the "primer" and "background" data. Finally, the training classification results are applied to actual detection. Before inputting the classification data "primer" and "background", the classification calculation formula and classification parameters are defined: the number of input layers, the number of hidden layer inputs, the classification calculation method, etc. When there is only one hidden layer, the linear formula can be obtained: h1=W1*X1+W2*X2+ W3*X3+W3*X3+b; where X represents the different features of the input layer image. When there are actually multiple hidden layers, h1=f(W[1,1]*X1+W[2,1]*X2+ W[3,1]*X3+W[4,1]*X4+b0)=f(A(1)n), where the function f(*) is the activation function and W represents the weight value. Add manually annotated samples to the classification formula. The number of selected samples will increase as the number of images increases. It's necessary to establish feature sets for the primer and background sample regions and add them together to the classification formula. Perform classification training based on the defined classification parameters and input variables. Set the training confidence level to determine if the training meets the requirements and obtain the trained model. If not, retrain by reselecting samples, increasing the sample size, adjusting training parameters, and so on.
[0030] Step 203, comparing the primer area with a preset detection template, and generating primer qualified information when all the primer areas match the preset detection template; After obtaining the primer area, all primer areas can be compared and matched with a preset detection template. Wherein, the preset detection template includes a coverage area, and comparing the primer area with the preset detection template includes: comparing the primer area with the coverage area; if the primer area and the coverage area are the same, determining that the primer area matches the preset detection template; if the primer area and the coverage area are different, determining that the primer area does not match the preset detection template.
[0031] In an embodiment of the present invention, the preset detection template includes at least a coverage area. The current primer area is compared with the coverage area. If the current primer area is identical to the coverage area, it indicates that the primer liquid in the current primer area can cover the desired area, and it can be determined that the current primer area matches the preset detection template. Conversely, if the current primer area is different from the coverage area, it indicates that the primer liquid in the current primer area does not cover the desired area, and it can be determined that the current primer area does not match the preset detection template.
[0032] If all the primer areas match the preset test template, the entire glass is qualified. The qualified primer can be referred to as Figure 3 , the primer liquid is uniform and complete as a whole, without drawing. At this time, the primer qualified information is generated to output the primer qualified information, and the primer qualified information is used to instruct relevant personnel to carry out subsequent processes.
[0033] Step 204: if at least one of the primer areas does not match the preset detection template, generate primer failure information.
[0034] If there is only one primer area that does not match the preset detection template, it means that there is a primer defect in at least one primer area, that is, the current primer process is not qualified. Figure 4 , there is a situation where the primer liquid is missing in at least one primer area. In this case, primer failure information can be generated. The primer failure information indicates that the current glass primer process is unqualified and the current glass needs to be repaired or scrapped.
[0035] In addition, in order to ensure the robustness and accuracy of the classification model, the currently collected glass primer image can be used for continuous training. The method further includes: saving the glass primer image to the annotated sample set.
[0036] The glass primer image is saved to the annotated sample set, and the updated annotated sample set is subsequently used for training until the corresponding convergence requirements are met to update the model and ensure the accuracy of classification.
[0037] The embodiment of the present invention collects a glass primer image during the glass primer coating process; performs regional classification on the glass primer image to determine at least one primer area; compares the primer area with a preset detection template, and generates primer qualified information when all the primer areas match the preset detection template; generates primer unqualified information when at least one primer area does not match the preset detection template. By collecting the glass primer image during the primer coating process and performing detection based on it, the primer condition of the glass can be detected in a timely manner, which improves the timeliness of the detection, so that primer defects can be repaired in a timely manner and improve production efficiency. The primer area and the background area are distinguished in the glass primer image, and then the actual glass primer condition is identified from the primer area based on the comparison with the preset detection template to determine whether the glass primer is qualified. This can reduce the situation where detection errors are caused by the glass and the background color being similar, thereby improving the accuracy of detection.
[0038] In order to make the implementation process of the embodiment of the present invention clear to those skilled in the art, the following reference is made to Figure 5 Let's illustrate with an example: 1. The parts (glass) arrive at the primer station and are ready to start primer coating. First, when the glass is transferred to the primer station, the primer coating process begins.
[0039] 2. The camera and brush head complete the primer along the preset trajectory, and the camera completes the image acquisition. During the primer process, the brush head first applies the primer along the preset trajectory. Simultaneously, the camera acquires the image to obtain the glass primer image.
[0040] 3. Using the pre-commissioning images, manually label the "primer" and "background" areas and trigger training. Before starting primer detection, the model needs to be trained using images collected during pre-commissioning. The model is trained using images labeled in red to enable it to identify primer and background areas. In other words, in an image, after the map area is identified, the remaining area is considered the background area.
[0041] 4. The training results are used as a detection template for primer inspection, and the detection results and the original image are output. The identified primer area is then tested for primer detection to determine whether it matches the detection template and obtain the detection results. The detection results and the original image are then output together.
[0042] 5. Use production images to continuously optimize the classification template based on the test results to achieve optimal detection. Use the tested production images again for model training, and continuously update the model to maintain its robustness and accuracy.
[0043] It should be noted that for the sake of simplicity, the method embodiments are described as a series of actions. However, those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.
[0044] Reference Figure 6 , shows a structural block diagram of an embodiment of a glass primer detection device of the present invention, which may specifically include the following modules: The first acquisition module 601 is used to collect the glass primer image during the glass primer coating process; A classification module 602 is configured to perform region classification on the glass primer image to determine at least one primer region; The comparison module 603 is used to compare the primer area with a preset detection template, and generate primer qualified information when all the primer areas match the preset detection template.
[0045] In an optional embodiment of the present invention, the device further comprises: In the case where at least one of the primer areas does not match the preset detection template, primer failure information is generated.
[0046] In an optional embodiment of the present invention, the device further comprises: The second acquisition module is used to obtain the labeled sample set and the initial model; A training module is used to train the initial model based on the labeled sample set until the training confidence of the initial model is greater than a preset confidence threshold, and generate a classification model, wherein the classification model is used to perform the step of regional classification of the glass primer image and determining at least one primer area.
[0047] In an optional embodiment of the present invention, the training module includes: The training submodule is used to input the labeled sample set into the initial model for training based on a preset hyperbolic tangent function.
[0048] In an optional embodiment of the present invention, the device further comprises: The glass primer image is saved in the annotated sample set.
[0049] In an optional embodiment of the present invention, the preset detection template includes a coverage area, and the comparison module 603 includes: a comparison submodule, for comparing the primer area with the cover area; a matching submodule, configured to determine that the primer area matches a preset detection template when the primer area is identical to the coverage area; The non-matching submodule is used to determine that the primer area does not match the preset detection template when the primer area is different from the coverage area.
[0050] The embodiment of the present invention collects a glass primer image during the glass primer coating process; performs regional classification on the glass primer image to determine at least one primer area; compares the primer area with a preset detection template, and generates primer qualification information when all the primer areas match the preset detection template. By collecting the glass primer image during the primer coating process and performing detection based on it, the primer condition of the glass can be detected in a timely manner, which improves the timeliness of the detection, so that primer defects can be repaired in a timely manner and improves production efficiency. The primer area and the background area are distinguished in the glass primer image, and then the actual glass primer condition is identified from the primer area based on the comparison with the preset detection template to determine whether the glass primer is qualified. This can reduce the situation where detection errors are caused by the glass and the background color being similar, thereby improving the accuracy of detection.
[0051] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
[0052] Reference Figure 7 The embodiment of the present invention further provides a glass primer detection device, comprising: a processor 701, a memory 702, and a computer program stored in the memory 702 and capable of running on the processor 701, wherein the computer program, when executed by the processor 701, implements the steps of the glass primer detection method described above. The glass primer detection method comprises: During the glass primer coating process, a glass primer coating image is collected; Performing region classification on the glass primer image to determine at least one primer region; The primer area is compared with a preset detection template, and primer qualified information is generated when all the primer areas match the preset detection template.
[0053] Optionally, the method further includes: In the case where at least one of the primer areas does not match the preset detection template, primer failure information is generated.
[0054] Optionally, the method further includes: Obtain annotated sample sets and initial models; The initial model is trained based on the labeled sample set until the training confidence of the initial model is greater than a preset confidence threshold, and a classification model is generated. The classification model is used to perform regional classification of the glass primer image and determine at least one primer area.
[0055] Optionally, the step of training the initial model based on the labeled sample set includes: The labeled sample set is input into the initial model for training based on a preset hyperbolic tangent function.
[0056] Optionally, the method further includes: The glass primer image is saved in the annotated sample set.
[0057] Optionally, the preset detection template includes a coverage area, and the comparing the primer area with the preset detection template includes: comparing the primer area to the covered area; In a case where the primer area is identical to the coverage area, determining that the primer area matches a preset detection template; In a case where the primer area is different from the covering area, it is determined that the primer area does not match the preset detection template.
[0058] The embodiment of the present invention collects a glass primer image during the glass primer coating process; performs regional classification on the glass primer image to determine at least one primer area; compares the primer area with a preset detection template, and generates primer qualification information when all the primer areas match the preset detection template. By collecting the glass primer image during the primer coating process and performing detection based on it, the primer condition of the glass can be detected in a timely manner, which improves the timeliness of the detection, so that primer defects can be repaired in a timely manner and improves production efficiency. The primer area and the background area are distinguished in the glass primer image, and then the actual glass primer condition is identified from the primer area based on the comparison with the preset detection template to determine whether the glass primer is qualified. This can reduce the situation where detection errors are caused by the glass and the background color being similar, thereby improving the accuracy of detection.
[0059] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor.
[0060] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.
[0061] Reference Figure 8 The embodiment of the present invention further provides a glass primer system, including a robot 801, a glass primer process equipment 802 and the glass primer detection equipment 803 as described above, wherein the glass primer process equipment 802 and the glass primer detection equipment 803 are installed on the robot 801; the robot 801 drives the glass primer process equipment 802 and the glass primer detection equipment 803 to move; the glass primer process equipment 802 is used to perform primer process treatment on the glass.
[0062] The robot 801 is the basis for performing the primer process and detection, and the glass primer process equipment 802 and the glass primer detection equipment 803 are installed on the robot 801. The glass primer process equipment 802 includes at least a primer liquid supply system and a primer brush head fixture. The primer brush head is installed on the primer brush head fixture, and the primer liquid supply system supplies primer liquid to the primer brush head. The robot 801 can move according to the production rhythm and teaching process. When the robot 801 moves, it drives the primer brush head fixture in the glass primer process equipment 802 to move, that is, the primer brush head moves accordingly, and the primer process is performed on the glass to be processed. At the same time, the glass primer detection equipment 803 will also move with the robot 801. After the primer brush head completes the brushing, the image after brushing can be collected, and the steps of the above-mentioned glass primer detection method can be executed to perform corresponding detection.
[0063] Specifically, the glass primer detection method includes: During the glass primer coating process, a glass primer coating image is collected; Performing region classification on the glass primer image to determine at least one primer region; The primer area is compared with a preset detection template, and primer qualified information is generated when all the primer areas match the preset detection template.
[0064] Optionally, the method further includes: In the case where at least one of the primer areas does not match the preset detection template, primer failure information is generated.
[0065] Optionally, the method further includes: Obtain annotated sample sets and initial models; The initial model is trained based on the labeled sample set until the training confidence of the initial model is greater than a preset confidence threshold, and a classification model is generated. The classification model is used to perform regional classification of the glass primer image and determine at least one primer area.
[0066] Optionally, the step of training the initial model based on the labeled sample set includes: The labeled sample set is input into the initial model for training based on a preset hyperbolic tangent function.
[0067] Optionally, the method further includes: The glass primer image is saved in the annotated sample set.
[0068] Optionally, the preset detection template includes a coverage area, and the comparing the primer area with the preset detection template includes: comparing the primer area to the covered area; In a case where the primer area is identical to the coverage area, determining that the primer area matches a preset detection template; In a case where the primer area is different from the covering area, it is determined that the primer area does not match the preset detection template.
[0069] The embodiment of the present invention collects a glass primer image during the glass primer coating process; performs regional classification on the glass primer image to determine at least one primer area; compares the primer area with a preset detection template, and generates primer qualification information when all the primer areas match the preset detection template. By collecting the glass primer image during the primer coating process and performing detection based on it, the primer condition of the glass can be detected in a timely manner, which improves the timeliness of the detection, so that primer defects can be repaired in a timely manner and improves production efficiency. The primer area and the background area are distinguished in the glass primer image, and then the actual glass primer condition is identified from the primer area based on the comparison with the preset detection template to determine whether the glass primer is qualified. This can reduce the situation where detection errors are caused by the glass and the background color being similar, thereby improving the accuracy of detection.
[0070] Reference Figure 9 The embodiment of the present invention further provides a computer-readable storage medium 901, wherein the storage medium 901 stores a computer program, and when the computer program is executed by a processor, the steps of the glass primer detection method as described in any one of the embodiments of the present invention are executed. The glass primer detection method includes: During the glass primer coating process, a glass primer coating image is collected; Performing region classification on the glass primer image to determine at least one primer region; The primer area is compared with a preset detection template, and primer qualified information is generated when all the primer areas match the preset detection template.
[0071] Optionally, the method further includes: In the case where at least one of the primer areas does not match the preset detection template, primer failure information is generated.
[0072] Optionally, the method further includes: Obtain annotated sample sets and initial models; The initial model is trained based on the labeled sample set until the training confidence of the initial model is greater than a preset confidence threshold, and a classification model is generated. The classification model is used to perform regional classification of the glass primer image and determine at least one primer area.
[0073] Optionally, the step of training the initial model based on the labeled sample set includes: The labeled sample set is input into the initial model for training based on a preset hyperbolic tangent function.
[0074] Optionally, the method further includes: The glass primer image is saved in the annotated sample set.
[0075] Optionally, the preset detection template includes a coverage area, and the comparing the primer area with the preset detection template includes: comparing the primer area to the covered area; In a case where the primer area is identical to the coverage area, determining that the primer area matches a preset detection template; In a case where the primer area is different from the covering area, it is determined that the primer area does not match the preset detection template.
[0076] The embodiment of the present invention collects a glass primer image during the glass primer coating process; performs regional classification on the glass primer image to determine at least one primer area; compares the primer area with a preset detection template, and generates primer qualification information when all the primer areas match the preset detection template. By collecting the glass primer image during the primer coating process and performing detection based on it, the primer condition of the glass can be detected in a timely manner, which improves the timeliness of the detection, so that primer defects can be repaired in a timely manner and improves production efficiency. The primer area and the background area are distinguished in the glass primer image, and then the actual glass primer condition is identified from the primer area based on the comparison with the preset detection template to determine whether the glass primer is qualified. This can reduce the situation where detection errors are caused by the glass and the background color being similar, thereby improving the accuracy of detection.
[0077] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.
[0078] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, apparatus, or computer program products. Thus, embodiments of the present invention may take the form of a fully hardware embodiment, a fully software embodiment, or an embodiment combining software and hardware. Furthermore, embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0079] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0080] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.
[0081] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.
[0082] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they become aware of the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.
[0083] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or terminal device that includes the element.
[0084] The above is a detailed introduction to a glass primer detection method, a glass primer detection device, a glass primer detection equipment, a glass primer system and a computer-readable storage medium provided by the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.
Claims
1. A glass primer detection method, characterized in that: include: During the glass primer coating process, a glass primer coating image is collected; Performing region classification on the glass primer image to determine at least one primer region; The primer area is compared with a preset detection template, and primer qualified information is generated when all the primer areas match the preset detection template.
2. The method according to claim 1, characterized in that The method further comprises: In the case where at least one of the primer areas does not match the preset detection template, primer failure information is generated.
3. The method according to claim 1, characterized in that The method further comprises: Obtain annotated sample sets and initial models; The initial model is trained based on the labeled sample set until the training confidence of the initial model is greater than a preset confidence threshold, and a classification model is generated. The classification model is used to perform regional classification of the glass primer image and determine at least one primer area.
4. The method according to claim 3, characterized in that The step of training the initial model based on the labeled sample set includes: The labeled sample set is input into the initial model for training based on a preset hyperbolic tangent function.
5. The method according to claim 3, characterized in that The method further comprises: The glass primer image is saved in the annotated sample set.
6. The method according to claim 1, wherein The preset detection template includes a coverage area, and the comparing the primer area with the preset detection template includes: comparing the primer area to the covered area; In a case where the primer area is identical to the coverage area, determining that the primer area matches a preset detection template; In a case where the primer area is different from the covering area, it is determined that the primer area does not match the preset detection template.
7. A glass primer detection device, characterized in that: include: A first acquisition module is used to collect a glass primer image during the glass primer coating process; a classification module, configured to perform regional classification on the glass primer image to determine at least one primer region; The comparison module is used to compare the primer area with a preset detection template, and generate primer qualified information when all the primer areas match the preset detection template.
8. A glass primer detection device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein when the computer program is executed by the processor, the steps of the glass primer detection method according to any one of claims 1 to 6 are implemented.
9. A glass primer system, characterized in that: include: A robot, a glass primer coating process equipment and a glass primer coating detection equipment as described in claim 8, wherein the glass primer coating process equipment and the glass primer coating detection equipment are installed on the robot; the robot drives the glass primer coating process equipment and the glass primer coating detection equipment to move; the glass primer coating process equipment is used to perform primer coating process on the glass.
10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the glass primer detection method according to any one of claims 1 to 6 are implemented.